知识卡片

普通读书笔记卡 · 3710.f.2.e

正半定二次型可近似解线性方程组

理解等级:中级

对对称正半定A,f(x)=½xᵀAx−bᵀx的梯度为Ax−b。

只有Ax=b可解(b在A的值域内)时,f才有极小点且极小点满足该方程;合适步长的梯度下降可逼近解。

若A=0且b≠0,目标无下界,不能称为近似求解。

这张卡片看不懂?

已有反馈 · 0 / 10 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • 第33章 机器学习算法 · 查看原书依据

      We now generalize this idea to higher dimensions, where using gradient descent may actually lead to a faster algorithm. One n-dimensional analog is the function , where A is an n × n matrix. The gradient of f with respect to x is the function Ax – b. To find the value of x that minimizes f, we set the gradient of f to 0 and solve for x. Solving Ax–b = 0 for x, we obtain x = A−1b, Thus, minimizing f(x) is equivalent to solving Ax = b. If f(x) is convex, then gradient descent can approximately compute this minimum. A 1-dimensional function is convex when its second derivative is positive. The equivalent definition for a multidimensional function is that it is convex when its Hessian matrix is

    参考来源

    - 位置:《算法导论(第4版)》第33章《机器学习算法》(OEBPS/chapter033.xhtml,文本字符[60000,65000))。 - 结论依据:第33.3节由∇f=Ax−b联系二次型与线性方程组,并要求 Hessian A 正半定以得到凸性;可解性仍是极小点存在的必要前提。 - 原始内容:We now generalize this idea to higher dimensions, where using gradient descent may actually lead to a faster algorithm. One n -dimensional analog is the function , where A is an n × n matrix. The gradient of f with respect to x is the function A x – b . To find the value of x that minimizes f , we set the gradient of f to 0 and solve for x . Solving A x – b = 0 for x , we obtain x = A −1 b , Thus, minimizing f ( x ) is equivalent to solving A x = b . If f ( x ) is convex, then gradient descent can approximately compute this minimum. A 1-dimensional function is convex when its second derivative is positive. The equivalent definition for a multidimensional function is that it is convex when its Hessian matrix is